Best of #econtwitter - Week of June 12, 2022 [2/2]
www.bestofecontwitter.com
Best of #econtwitter - Week of June 12, 2022 [2/2]
Welcome readers old and new to this week’s edition of Best of Econtwitter. Thanks to those sharing suggestions, over email or on Twitter @just_economics.
This is part two of two.
Paper summaries
Economists love using linear regression to estimate treatment effects — it turns out that there are perils to this method, but also amazing perks
Come with me in this 🧵 if you want to learn…
Pedro H. C. Sant'Anna @pedrohcgs
^model of a paper summary thread; another thread from a coauthor:
Why, you might ask? Well for one thing @borusyak and I show in our recentered IV paper that regression adjustment for the propensity score / expected instrument guarantees convex average TE estimation for a single treatment in a giant class of settings
dropbox.com/s/a40ksnsi0geb…
Peter Hull @instrumenthull
@arindube Regressions are great when you have a single treatment or IV with a clear design. I hope we can get that message out
^and a blog post summary, with simulations. “First they came for our DiD/TWFE, now they're coming for our OLS. Econometricians must be stopped.”
Graduate degrees vs. satisfaction with job and contribution to society.
MDs have the highest satisfaction.
MBAs — not so much.
Source: NBER paper by Altonji, Humphries & Zhong
nber.org/papers/w30105
We’re pleased to announce the release of our new WP "Wealth of two nations: The US racial wealth gap, 1860-2020," in which we provide the first continuous time series of white-to-Black per capita wealth ratios over the past 160 years. 1/10
nber.org/papers/w30101
🚨NEW PAPER🚨 We examine high-stakes strategic choice using more than 40 (!) years of data from the American TV game show "The Price Is Right". Want to know what we find? Read the thread below!👇
[or the paper here: papers.ssrn.com/abstract_id=41…] #EconTwitter
Random exposure to higher likelihood of being prescribed opioids leads workers to have worse cognitive performance and more behavioral issues.
^“The US military has great data if you can get access to it, which I cannot”
More: SG payroll RDD; digitalization enhances tax compliance; mentorship to Saudi Arabian girls; forced sterilization; eviction trial commutes/more; globalization meta-analysis; “Evolution, uncertainty, and the asymptotic efficiency of policy”; female headteachers
Interesting discussions
During my economic PhD, I must admit I had no idea what people were talking about. It’s not that I didn’t understand the answers. I couldn’t understand the questions and why they were even questions. I understood nothing in the seminars for years until I just accepted it. 1/4
I always assumed it was bc I hadn’t spent enough time reading the literature to understand why the questions evolved to be the way they were. It wasn’t until well after my PhD did I pick up the confidence to realize that whatever didn’t make sense really didn’t. 2/4
Suppose a risky action and a safe action, with the Sender more cautious than the Receiver.
The optimal persuasion scheme pools high-risk states, until the Receiver is indifferent about whether to follow the advice.
The FDA and CDC advise that you don’t eat raw cookie dough.
“I don’t want to go to grad school or become a professional philosopher, but I want more philosophy in my life. How can I get this?”
This is the question I hear most. Please help me come up with a GREAT answer to it. Leave your thoughts in the replies so I can steal them.
^if someone tweets this for econ… tag @just_economics
The winner's curse has a possibly interesting application for the experience of people in high-risk careers, such as academia or professional sports.
You can think of such a career as an all-pay auction: everyone tries, only some get a prize.
1/
make an economist cry with just 4 words
Adi Polak @AdiPolak
